UX Research Methods: How Many People Each One Needs in India

Match the method to the kind of answer you need. Interviews, diary studies and field visits show why people behave as they do. Usability tests show whether a design works. Surveys, A/B tests and analytics count how many people do something. Early in a product, learn why. Later, test and measure.

Two ideas keep this straight. An attitudinal method records what people say; a behavioural method records what they do. Interviews are attitudinal, so they capture reported behaviour, not observed behaviour (Nielsen Norman Group, 2023). If a decision depends on what people actually do, pair interviews with a usability test or a field visit.

The product stage narrows the choice further (Nielsen Norman Group, Rohrer, 2022, reviewed July 2026):

• Strategize: field studies, interviews and concept testing (checking an idea with users before it is built).
• Design: usability testing, card sorting (people group topics into categories) and tree testing (people try to find items in a text-only menu).
• Launch and assess: benchmarking (measuring a product against a baseline), A/B testing and analytics.

Each method has a starting number and a trigger to raise it, set out in the sheet below. Raise the number only when your users differ from one another, and size surveys and A/B tests by the precision you need.

Most method lists name methods but not participant numbers or Indian recruitment constraints.

Four notes on reading the sheet:

• Saturation: the point where more interviews stop changing your themes. Start small, analyse as you go, and stop there.
• Usability testing: spend extra budget on more rounds, not more users per round.
• Diary studies: examples run from one to three weeks. Recruit extra people because some drop out, and pay more than you would for a single usability session (Nielsen Norman Group, 2024).
• Choosing a usability format: our remote usability testing guide covers moderated against unmoderated (participants do it alone) formats. For the wider reasoning on qualitative samples, see Sample Size in Qualitative Research.

Recruiting only through online, urban channels can skew an Indian research sample. As of June 2026, India had 129.03 internet subscriptions per 100 people in urban areas and 49.07 in rural areas. Urban density is about 2.6 times rural. That is the telecom regulator's own count for April to June 2026.

Rural areas hold 448.08 million of 1,114.17 million internet subscriptions, which is 40.2%. We worked out the 2.6 and the 40.2% by dividing the Telecom Regulatory Authority of India's figures. Both count subscriptions, not unique people, so a figure above 100 per 100 means some people hold more than one connection (our inference).

The data does not measure how a sample skews. It does show where connections are concentrated. Our inference is that a remote-only study recruited through metro channels will under-represent a large share of connections. If your product serves tier-2, tier-3 or rural users, put at least one segment outside the metros and test on the phones and connections those users have.

Language works the same way. Treat each group you expect to behave differently, such as a city and language pair, as its own segment. Diverse samples sit at the top of the 20 to 30 or more interview range, not the bottom.

Work backwards from completed sessions. Divide by the share who show up to get bookings, then divide by incidence (the share of screened people who qualify) to get people to screen. The example below is our model with assumed rates, not an official figure.

Illustrative scenario. A team building an app on UPI (Unified Payments Interface) wants to understand first-time payers, then test a redesign.

• Interview segments: 3 cities × 2 languages = 6 cells.
• Interviews: budget 5 per cell, so 6 × 5 = 30. This sits inside the 20 to 30 or more range for varied users. Run in waves and stop early if later waves repeat earlier themes. At 60 minutes each, that is 30 hours of interviewing.
• Usability tests: we keep the redesign test to one chosen segment (one city, one language), so 2 rounds × 5 users = 10 sessions. If you test more than one distinct user group, use 3 to 4 users per group per round (Nielsen Norman Group, 2012) and recompute the lines below.
• Completed sessions: 30 + 10 = 40.
• Assumed show-up rate: 80%. Bookings needed: 40 ÷ 0.8 = 50.
• Assumed incidence: 20%. People to screen: 50 ÷ 0.2 = 250.

Swap in your own rates. If only 1 in 10 people qualifies, the same 50 bookings need 500 screened. Our incidence rate guide shows how to estimate the rate before you commit.

Run these five checks, each with a clear pass or fail, before the first session.

You send us a brief, we confirm the study is feasible, and we recruit, schedule and run it. Your brief covers the research objective, method, target audience, participant profile, city, sample size, usership, timeline and recruitment criteria. We then send a tailored proposal and quote.

We recruit for and run these formats:

• In-depth interviews: one-to-one, 30 to 120 minutes, online or in person.
• Focus groups: moderated, 1 to 3 hours, online or in person.
• Digital diary studies: multi-day photo and video tasks on a mobile app or the web.
• Ethnography: home visits or virtual sessions.
• Usability and app testing: remote or lab-based.
• Also available: online bulletin boards, shop-alongs, in-home use tests and opt-in digital behavioural tracking.

For the recruit, we can recruit by city, target by household and demographic profile, run sessions in languages other than English, supply moderators, and check participant identity before invitation. Studies may be run by us, by research partners or by other research organisations, depending on the project. Our approach is guided by the ethical principles of the Market Research Society of India and ESOMAR.

We recruit participants from our panel. People who want to take part in studies can register their interest through how to join and get invited; registration does not guarantee invitations, which depend on each study's requirements and the person's profile.

What you receive depends on the scope we agree before the study starts. It can include recruitment, fieldwork materials, recordings, transcripts, structured data exports and reports. We price each study as a project, with no fixed rate card, and we offer a free pilot or sample.

To start, email your brief to the Qualfacto team or message us on WhatsApp. We aim to reply within one business day (Monday to Friday, 10:00 to 18:00 India Standard Time); detailed briefs may take longer.